← authors · overview

• Vinvl: Revisiting Visual Representations In Vision-language Models
  2021 · 862 citations
• Unified Vision-language Pre-training For Image Captioning And VQA
  2019 · 836 citations
• Regionclip: Region-based Language-image Pretraining
  2021 · 495 citations
• Multi-scale Vision Longformer: A New Vision Transformer For High-resolution Image Encoding
  2021 · 322 citations